PREVENTIVE TREATMENT AND CONTROL TECHNIQUES … · preventive treatment and control techniques for...

A. Taylor et al.484

Treatment Measures

PREVENTIVE TREATMENT AND CONTROL TECHNIQUESFOR BALLAST WATER

ALAN TAYLOR1*, GEOFF RIGBY2, STEPHAN GOLLASCH3,MATTHIAS VOIGT4, GUSTAAF HALLEGRAEFF5,TRACY MCCOLLIN6 & ANDERS JELMERT7

1Alan H. Taylor & Associates Melbourne, Victoria, Australia2Reninna Consulting Pty. Limited, Charlestown N.S.W, Australia3GoConsult, Hamburg, Germany4 Dr. Voigt-Consulting, Stolpe, Germany5University of Tasmania, Hobart, Tasmania, Australia6FRS Marine Laboratory Aberdeen, Aberdeen, Scotland7Floedevigen Research Station, Norway*Corresponding author [email protected]

AbstractIt is apparent that no single or simple universal solution presently exists for shipboard treatmentor management to prevent the transfer of viable non-native organisms in ballast water. Only avery limited number of the treatment options listed below has been shown to be 100% effective(and only for some specific organisms in some cases), environmentally sound, cost effective andsafe during application. Several technologies (including the widely recognised forms of waterexchange currently in use throughout the world) or a combination of technologies (tool box) mayhowever, be at least partially effective and feasible in terms of economic and shipboard con-straints. Currently heat treatment, mechanical removal of organisms in combination with UVtreatment, and chemical treatment of ballast water are considered the most promising approaches.However, concerns have been expressed regarding residual environmental polluting components,health and safety problems related to storage of chemicals and compatibility with cargo carried onboard as well as direct and indirect handling of chemicals by crew members. Given the globalnature of shipping and therefore the transport of non-native organisms in ballast water, the Inter-national Maritime Organization (IMO) and various countries, are considering the adoption of aBallast Water Convention, which would include a technical effectiveness standard that wouldform the initial basis for acceptance of the various treatment options. The IMO Convention wouldaim to achieve a standard approach to ballast water management but in the meantime some coun-tries, such as Australia, have introduced their own ballast water regulations. It is important thatongoing research and development aimed at developing and demonstrating existing and newtreatment techniques be maintained in order that the IMO can incorporate the best available tech-nology into the proposed convention. In summary, there is no current “stand-alone” treatmentoption, that covers all possible scenarios. However, a combination of methods could result incost-effective management options.

mailto:[email protected]

Treatment and control of ballast water 485

1 Introduction

The amount of ballast water transported annually on a world-wide basis has been esti-mated by scientists and engineers to be around 10 billion tonnes (Rigby & Taylor 1995)highlighting the global dimension of the problem to be controlled. It was further esti-mated that some 7,000 taxa are being transported in international shipping each day(Carlton pers. comm.). One reason for this great diversity of organisms arises from thethree different “habitats” inside ballast water tanks: (i) tank walls, (ii) ballast water, and(iii) the sediment. Because of the diversity in ship design and improved technology (e.g.double hulls, higher economical cruising speeds), the survival rates of some speciescarried in ballast tanks have increased, and consequently, many introductions of nonin-digenous organisms in new locations have occurred in recent years highlighting theneed for effective ballast water management.

The control of ballast water introductions is likely to be based on a quarantine approach.This approach does not intend to provide an absolute barrier to prevent the introductionof unwanted nonindigenous species (Carlton et al. 1995) but rather aims for a signifi-cant reduction in risk. It has become clear that no single treatment process is likely touniversally achieve the required inactivation, kill or total removal of all unwanted or-ganisms in ballast water. Although a single process may be appropriate in some circum-stances, a two stage treatment approach may be necessary to achieve the required levelof effectiveness in other cases. Two stage treatment may comprise some form of me-chanical removal of organisms followed by a physical or chemical treatment method.Comprehensive overviews on physical and chemical treatment options have been com-piled, however, the depth of the information available on the individual treatment op-tions varies greatly. While some techniques have been tested in the laboratory only withsingle or a few species, others have followed a long and costly test procedure involvingnumerous species and/or large scale trails in land based facilities or onboard ships (Fig.1).

The mention of trade names and producers in this contribution does not necessarilyimply an endorsement of the product by the authors but is mentioned as an example ofthe relevant technology.

2 Research initiatives on ballast water treatment

Since 1995 some fifteen research initiatives on ballast water treatment have been com-pleted. Project objectives included the testing of several treatment options with regard totheir effectiveness, environmentally soundness and practicability on board of vessels.The treatment options considered include ballast water exchange, filtration, heat treat-ment, coagulation/flocculation, pH adjustment, chemical treatment (ozone, glutaralde-hyde, oxygen deprivation, hydrogen peroxide based formulations and chlorine) and UV.A further 14 projects with comparable objectives are underway and will be completedby 2003 (GloBallast Programme 2001). At this stage the various methods of treatmentthat have been put forward have been frequently referred to as a "tool box" from whichthe most effective (in achieving the desired level of effectiveness), practical (safe andeasy to apply, not compromising deck safety or operational requirements, not damagingexisting ship installations such as pumps, ballast tank integrity, ballast tank coating,

A. Taylor et al.486

isolators and sealing rings), cost effective and environmentally sound combinationshould be selected. Several of the options are straightforward statements of best practicebut in many circumstances the choices available to a ship's Master may be limited.

Figure 1. Port based and shipboard ballast water treatment options.

3 IMO Assembly Resolution

To date, international guidelines have been adopted as the IMO Assembly ResolutionA.868(20) (see below). The IMO has not generally promoted regionally different sys-tems, emphasising that a universal global approach is preferred to solve the ballast waterproblem. However it has been realised that some local restrictions may be appropriate tomanage or control a particular organism of concern. Further, using different manage-ment options and treatment techniques could result in unwanted regional restrictivepractices, restraints of trade and competitive advantages. Unfortunately, some con-cerned countries have already implemented voluntary and mandatory guidelines onballast water, mainly (at least initially) based on the concept of ballast water exchange.

The International Maritime Organization’s Marine Environment Protection Committee(MEPC) has had a specific interest in the field of unwanted aquatic species introduc-tions by ballast water since 1973 when the International Conference on Marine Pollu-

Ballast water treatment technologies

Port based Shipboard

Ballast pre-treated water

Treatment afterdeballastingLand based plantReceiving vessel Mechanical re-

movalFiltrationCyclonic separationSedimentation/flotationHigh-speed pumpsOthers

Physical inactiva-tionHeatingCoolingUV IrradiationUltrasonicMicrowaveElectrical inactivationGas supersaturationOthers

Ballast waterexchange

Empty/refillContinuous flow-throughDilution method

Chemical and bio-cidal inactivationHypochloriteClO2OzonePeracetic AcidHydrogen PeroxideAldehydesQuinolonesOthers


tion adopted Resolution 18, drawing attention to the transport of aquatic organisms andpathogens around the world in ships´ ballast tanks. In the late 1980s, the MEPC formeda working group to consider research information and solutions proposed by MemberStates of the IMO and by non-governmental organizations. The working group con-cluded that voluntary guidelines were the appropriate first step in addressing this prob-lem. MEPC adopted guidelines by resolution in 1991 and in 1993 these were adopted bythe IMO Assembly under Resolution A.774 (18) entitled "International Guidelines forPreventing the Introduction of Unwanted Aquatic Organisms and Pathogens from ShipsBallast Water and Sediment Discharges". This was then replaced in 1997 by the IMOAssembly Resolution A.868 (20) "Guidelines for the Control and Management ofShip’s Ballast Water to Minimize the Transfer of Harmful Aquatic Organisms andPathogens". The IMO Assembly Resolution A.868 (20) includes a recommendation thatan exchange of ballast water is carried out in open oceans as far as possible from theshore. The mid-ocean exchange of ballast water is currently the only readily availablemethod that can be used in order to minimize the risk of transfer of unwanted organismson existing vessels. Compared with coastal waters, deep ocean waters are generallyexpected to contain fewer organisms and, in addition, species occurring in open oceanwaters are generally not able to survive in coastal zones and vice versa. Where openocean exchange is not possible, requirements developed within regional agreementsmay be applicable, particularly in areas within 200 nautical miles from shore.

It is noted that no form of ballast exchange should be undertaken unless it is included inthe ship's Ballast Water Management Plan and approved by the ship’s ClassificationSociety via the ship’s “Trim and Stability” booklet. It is always the responsibility of theship's Master to ensure that any operation carried out at sea is done so in a safe manner.In addition to the exchange of ballast water at sea the guidelines include reference toballast water management practices that would reduce the risk of introducing non-nativespecies such as:(i) ballast water uptake should be avoided in the presence of harmful algal blooms and

known unwanted contaminants (e.g. Cholera disease outbreaks),(ii) precautionary procedures when taking on ballast water in shallow areas,(iii) discharging ballast water and sediments to on-shore facilities (if available) and(iv) avoiding ballast water uptakes at night as many zooplankton organisms migrate

towards the water surface in darkness.

The MEPC Ballast Water Working Group is currently working towards a set of legallybinding regulations in the form of a stand alone IMO Convention. The working group isscheduled to present the final draft of the guidelines to the Committee in 2002 and adiplomatic conference for its adoption is planned for 2003.

The following list of management options and treatment techniques to reduce the risk ofspecies translocations (not necessarily in order of preference or effectiveness) providesan overview of options and does not claim to be fully comprehensive.

3.1 BALLAST WATER EXCHANGE

Ballast water exchange was originally developed as a method to be used by vessels ontrans-oceanic journeys. The basis is that the water that was loaded in the port is ex-

A. Taylor et al.488

changed for deep oceanic water that will contain fewer organisms that are unlikely tosurvive in coastal waters. The exchange process was also recommended for when avessel was travelling between two fresh water ports as the increase in salinity would killany freshwater organisms remaining in the tanks and the oceanic species would notsurvive in freshwaters. There are currently three methods by which a mid-ocean ex-change of ballast water may be achieved:

3.1.1 Option 1 Empty/refill (=reballasting)An empty-refill is as it sounds; the ballast tanks are emptied of port water and thenrefilled with oceanic water. Stripping pumps or eductors should be used wherever pos-sible to minimise the amount of originally ballasted water remaining in the tanks. Trialshave shown that for deballasting undertaken in such a manner at least 95% of the origi-nal water can be replaced (Table 1) (Rigby 1994; Rigby & Taylor 2001; Miller 1998;Wonham et al. 1996). However, it has to be noted that the exchange of 95% of the vol-ume of the ballast water may not be equivalent to the exchange of 95% of the organismsin ballast tanks as these are not necessarily equally distributed in the ballast water, butmay accumulate at the bottom and tank walls (GloBallast Programme 2001). On manyships this method may result in unacceptable bending moments or shear stresses (Rigby& Hallegraeff 1994; Karaminas 2000), but potentially could be 100% effective at re-moving all the original ballast water on some vessels. In practice, many woodchip carri-ers that claimed to have undergone reballasting still had sediments present in the tanksthat included toxic dinoflagellate cysts (Hallegraeff & Bolch 1991, 1992). (Details onsafety aspects further below).

3.1.2 Option 2 Continuous flow-through of ballast water (=ballast exchange)A continuous flow through system allows continuous sea-to-sea circulation of ballastwater while the ballast tanks remain filled, i.e. sea water is pumped into the ballast tankswhile the tank is simultaneously overflowed from the top of the tank. Where the flow-through method is employed, at least three times the tank volume should be pumpedthrough the tank (on some vessels this has been shown to correspond to a replacementof approximately 95% of the original water). Some pipework modifications may benecessary on some ships to enable this option to be utilised safely and effectively (Tay-lor & Rigby 2001).

In contrast to deballasting in high seas during bad weather using the empty/refill tech-nique, the continuous flow through system does not impose excessive bending momentsor shearing forces and minimises stability problems. However, Rigby & Hallegraeff(1993, 1994) demonstrated that by emptying certain ballast tanks on the bulk carrierIron Whyalla the still water bending moment may be much higher than the maximumallowable value. This fact, in combination with the high number of organisms in theremaining water bodies in the ballast tank after emptying (option 1, see above), madethe flow through option more favourable. However, future research should be carriedout to confirm this view which is based on a limited number of trials.

3.1.3 Option 3 Dilution MethodThe dilution method is a further development of the continuous flow through technique.After the installation of additional pipework on the vessel continuous ballasting from


the top of the tanks via one pipe system and at the same time continuous deballasting bya second pipe system at the bottom of the tank may be carried out (IMO MEPC 38/13/21996; Villac et al. 2000). Mathematical modelling of the effectiveness of this methodwas carried out resulting in a comparable effectiveness as the ballast tank flushing forthree times the ballast tank volume (Armstrong et al. 1999).

3.2 EFFECTIVENESS OF THE BALLAST WATER EXCHANGE

3.2.1 Water replacement efficiencyThe efficiency of replacement of the original water will depend on the design of theship's ballast tank, safety requirements, sea conditions, quantity of water pumped andthe pumping system design. Water replacement efficiencies have recently been re-viewed by Rigby & Taylor 2001 and Rigby 2001 (Table 1). Trials have shown thatthree times volumetric exchange of ballast water result in approximately 95% removalof viable algal cells and approximately 60% removal of zooplankton organisms. How-ever, as the inoculation size to introduce a new algal species is about 1,000 cells a re-moval of 95% of these organisms would not eliminate the risk of species invasions asmillions of specimens remain in the water (GloBallast Programme 2001). The efficiencyof water exchange using the continuous flushing option is lower if the ship is not at sea,as the mixing is less efficient compared to when the vessel is moving as at sea (Figure2; Rigby & Hallegraeff 1994).

Table 1. Efficiency of water exchange for various ocean exchange options.Mode of Exchange % Water

exchangedReference

Continuous flushing, 0.5 tank volume 39.3 Rigby & Hallegraeff 1994Continuous flushing, 1 tank volume 63.2 Rigby & Hallegraeff 1994Continuous flushing, 2 tank volumes 86.5 Rigby & Hallegraeff 1994Continuous flushing, 3 tank volumes 95.0 Rigby & Hallegraeff 1994Continuous flushing, 4 tank volumes 98.2 Rigby & Hallegraeff 1994Continuous flushing, 3 tank volumes >90, 99 Taylor & Bruce 2000Empty/refill (20 tonnes residual per db tank) 99.6 Calculated(calculated for 50 tonnes residual water) 99.2Non-dedicated tanks empty/refill 100 Rigby 1994Empty/refill 95.0 Miller 1998Dilution/flushing 90.0 IMO MEPC 38/13/2 1996,

1998Dilution/flushing 86-96 Villac et al. 2001Sequential empty/refill >99 Wonham et al. 1996Continuous flushing/heating >99 Rigby et al. 1997Ocean exchange for salinity increase ofbrackish water;(ocean 35 PSU, brackish 15 PSU) 75(ocean 35 PSU, brackish 5 PSU) 83

3.2.2 Organism removal efficiencyThe efficiency of removal of organisms as distinct from the original ballast water, is acomplex issue, which will be affected by the nature and behaviour of organism in thetanks, the design of tanks, mixing within the tanks and the types and behaviour of sedi-ments (Table 2). For example some fast swimming zooplankton may remain after sev-

A. Taylor et al.490

eral tank volumes have been replaced, as indicated by recent sampling studies on theIron Whyalla (Sutton et al. 1998).

Figure 2. Continuous flushing ballast exchange efficiencies on the Iron Whyalla (Rigby &Hallegraeff 1994).

Shipboard based microscopic examination of organism removal during ballast exchangeand heating trials on the Iron Whyalla, showed that in addition to the effects of heating,the efficiency of removal of phytoplankton contained in the water after heat treatmentwas similar or higher than for water exchange (Rigby & Hallegraeff, 1994; Rigby et al.1997). Flushing trials by Ruiz & Hines (1997) in wing tanks on the S/R Long Beach andthe S/R Benecia showed a 60% water exchange on the basis of salinity and less than90% on the basis of coastal plankton communities. In another trial, the exchange of 3tank volumes resulted in water exchange efficiencies of 70-100% on the basis of salinityand greater than 95% on the basis of coastal organisms.

In a recent trial involving the Dilution Method of continuous flushing on the oil carrierM/V Lavras, a water exchange efficiency of 90% was achieved with a phytoplanktonexchange of 96%. Chlorophyll ‘a’ exchange was estimated as 86% (Villac et al. 2001).

From the flow through ballast exchange trials carried out by Taylor & Bruce (2000), itwas concluded that there was retention of phyto- and zooplankton from the source portfor the Spirit of Vision trials and that biological efficacy was at variance with the tracerdilution efficiencies. Approximately 50% of one of the taxa, the diatom Thalassiosiraspp., appeared to settle in the bottom layer of the tank after exchange. However, thetrials on another vessel, the Iver Stream, indicated that the flow through dilution methodwas relatively effective at reducing the planktonic organisms originally ballasted in thesource port. A 90-100% reduction in the means of depth stratified counts of source portindicator taxa was achieved.


Many other studies showing survival of organisms have been reported following ballastexchange in some form or other, however in many cases there has been no quantitativeassessment of replacement of the originally ballasted water and researchers have reliedon the word-of-mouth comments of the ship’s officers. Consequently, it is difficult toaccurately compare differences between water exchange and biological exchange. Forexample, Locke et al. (1993), using freshwater zooplankton and salinity as indicators(for ships originating from fresh and brackish water ports), estimated that for 24 vesselsentering the Great Lakes region, the efficiency of zooplankton exchange was 67% andfor water exchange was 86%.

Table 2. Efficiency of organism removal for various ocean exchange options.Mode of Exchange % organisms removal Reference

continuous flushing >95 (phytoplankton) Rigby & Hallegraeff 1994;Rigby et al. 1997

continuous flushing <90 (coastal organisms) Ruiz & Hines 1997empty/refill, 3 tank volumes >95 (coastal organisms) Ruiz & Hines 1997continuous flushing 96 (phytoplankton) IMO MEPC 1998flow through 90-100 (selected taxa) Taylor & Bruce 2000empty/refill, 1 tank volume 67 (plankton), estimated Locke et al. 1993empty/refill, 1 tank volume 48 (phytoplankton) Dickman & Zhang 1999dilution 86-96 (phytoplankton) Villac et al. 2001

Harvey et al. (1999) reported that although complete exchanges had been reported forvessels entering the Great Lakes, the presence of coastal species suggested that theexchanges were incomplete.

During a scientific study in the framework of the European Concerted Action”Introductions with ships” (1997-1999) the container vessel Pusan Senator was accom-panied in 1999 on a voyage from Kaohsiung (Taiwan) to Hamburg (Germany) to inves-tigate the zooplankton present in ballast tanks over time (Fig. 3). In addition, the ballastwater was exchanged enroute using (three times) the empty/refill method. A comparisonof the zooplankton community before and after exchange revealed that the number ofspecies increased considerably after water exchange but in contrast the number of indi-viduals showed a continuously decreasing trend. Some species survived the entire 23day voyage. It was concluded that a mid-ocean exchange does not necessarily result in acomplete exchange of the taxa in the ballast water and reduces the density of zooplank-ton individuals inside the tank on a limited scale.

In examining the effectiveness of open ocean reballasting (empty and refill) in reducingthe number of diatoms and dinoflagellates in ballast water for 14 container ships travel-ling from Oakland, California to Hong Kong, Zhang and Dickman (1999) reported theMaster estimated that 95 to 99% of the original water was removed and their analysesshowed that 87% of the diatoms and harmful dinoflagellates had been removed. In asimilar study (Dickman & Zhang 1999) for 5 container ships travelling from Manza-nillo, mid ocean reballasting resulted in 48% removal of diatoms and dinoflagellates(again with the Master estimating 95-99% water replacement). The differences betweenthe two studies was attributed to the fact that the latter study involved much older ships,which did not have such efficient ballast water exchange systems whereas the first studyinvolved ships that had only been in use for approximately 1 year and had more modern

A. Taylor et al.492

ballast water exchange designs. However in the absence of any quantitative measure-ments on water exchange it is difficult to make a direct comparison.

02000400060008000

100001200014000160001800020000

21.05

.1999

24.05

.1999

26.05

.1999

28.05

.1999

30.05

.1999

01.06

.1999

03.06

.1999

06.06

.1999

08.06

.1999

10.06

.1999

12.06

.1999

14.06

.1999

Spec

imen

s

0

5

10

15

20

25

30

35

Taxa

SpecimensTaxa

[n/100L] [n]

Mid-ocean exchange of ballast water carried out

May 30th.

Figure 3. Number of zooplankton taxa and individuals from the ballast water of the port-side side tank during a ships voyage from Kaohsiung to Hamburg. Origin of the ballastwater investigated: Hong Kong. After sampling on May 30th a mid-ocean exchange of theballast water was undertaken. All samples were taken using plankton net with a meshsize of 55 µm.

The above limited observations and the diversity of results concerning differences be-tween water exchange or replacement efficiency and removal/replacement of organismsshow that it is not considered realistic to seek to establish a generalised definitive rela-tionship between the two. In addition to natural mortality in the tanks, tank geometryand ballast hydraulic design, type of exchange operation and other operational featuresspecific to each ship and the types of organisms present will have a marked effect ontheir behaviour within the tanks.

3.3 SIGNIFICANCE OF SEDIMENTS

The significance and role of sediments in relation to ballast exchange has been poorlyresearched. Analyses of sediments removed from tanks have identified the presence ofharmful organisms. For example, Hallegraeff & Bolch (1992) have reported the pres-ence of toxic dinoflagellate resting spores in various types of sediments. It is feasible forsediments to remain in ships for many years (interdocking interval). However, in manyships, sediment is confined to areas well away from locations where the main flow ofwater occurs and hence may not play a significant role on organism dispersal. Workcurrently in progress is aimed at examining some of these effects in more detail (Rigby& Hallegraeff 2001).


3.4 LIMITATIONS OF BALLAST EXCHANGE

Whilst ballast exchange is currently the primary option available to owners, operatorsand ship’s Masters for treatment of ballast water, there are some limitations and poten-tial disadvantages as far as the possibility of organism translocation is concerned.

The precise location of ballast water exchange needs to be carefully chosen. In studiesby Macdonald & Davidson (1998) and Forbes & Hallegraeff (2001) mid ocean ex-change of ballast water resulted in an increase in the species diversity of phytoplanktonin ballast tanks. Macdonald & Davidson (1998) found that although exchange of ballastwater in regional seas may reduce the risk from polluted European harbour waters, itmay result in transport of potentially harmful phytoplankton species from the North Seaor Irish Sea to UK coastal areas, where conditions are likely to be sufficiently similarfor survival. Details of the type of mid ocean exchange were not reported in detail inthis study, but it was reported that the diversity of diatoms and dinoflagellates increasedin 69% and 85% of cases, and abundance increased in 31% and 85% of cases respec-tively. A more detailed follow on study using one of the vessels from the preliminarystudy is being carried out by McCollin et al. (2001). This study involves taking samplesbefore, during and after an exchange process. The vessel can carry out either an empty-refill or a flow through exchange process. The project is currently in the initial stagesbut preliminary results would seem to indicate that the exchange process in regionalseas is not as effective as in deep oceanic waters.

In a recent study Forbes & Hallegraeff (2001) reported that 80% of woodchip shipsoperating between Japan and Triabunna in Tasmania reballast in coastal waters offPapua New Guinea and bring a new viable tropical/cosmopolitan inoculum, mixed withremnants of old Japanese plankton into Triabunna. Monitoring of woodchip vesselsarriving in Triabunna, which claim to have exchanged 100%, indicated that 80% ofships still contain up to 30 culturable diatom species (including potentially toxicPseudo-nitzschia).

Forbes & Hallegraeff (2001) noted that there may be the problem of ships carryingtropical plankton directly into North Australian tropical ports (such as Port Hedland andGladstone), without undertaking any exchange and relying on the natural ballast watermortality to take its toll. Here it would be preferable to exchange the water. Scientificstudies have shown that even after 116 days living zooplankton can be found in ballasttanks (Gollasch 1996) and under certain circumstances zooplankton species reproducein ballast tanks (Lenz et al. 2000). Further Forbes & Hallegraeff (2001) showed that theexchange of ballast water could increase the species diversity of ballast tanks, especiallyin many domestic shipping routes, where no deep water exchange zones occur.

These observations illustrate that although mid ocean exchange is currently the mostwidely practised treatment option, caution needs to be exercised in assuming that use ofthis option will always reduce the risk of translocating undesirable organisms, espe-cially if the exchange water is compatible with the receiving water. In utilising ex-change on relatively long international voyages, care will need to be exercised in thefuture in choosing areas for exchange where organisms in the ocean water are not likelyto be more of a problem than those present in the original ballast water. Similar cautionneeds to be taken for shorter international and coastal voyages where suitable rebal-

A. Taylor et al.494

lasting zones may be limited or insufficient time available for exchange in an appropri-ate area.

It is also relevant to re-iterate that exchange of water has been adopted as an initialattempt to minimise the risks of discharging harmful organisms in ballast water and thatresearch aimed at developing and demonstrating more effective treatment optionsshould be encouraged and continued, so that improved systems can be utilised in thefuture.

Even if it is assumed that the efficiency of removal of organisms is the same as thewater replacement efficiency in ocean exchange, it is important to realise that largenumbers of harmful organisms may still be present in the water discharged into thereceiving port. This is especially true when ballasting occurs during an algal bloom inthe ballasting region. This point can be illustrated by considering an algal bloom of thetoxic dinoflagellate, Gymnodinium catenatum. Typical cell densities during a bloommight be 105/L. This means that in a Cape Size carrier containing 55,000 m3 ballastwater, the number of cells present would be 5.5x1012. Typically, it can be assumed thatabout 1% of these cells could produce cysts, and hence the number of residual cysts inthe water remaining on the ship after flushing with three tank volumes (assuming anexchange efficiency of 95%) would be 2.75x109. This number of cysts would obviouslyrepresent a very serious threat when discharged into a receiving port, since a viableinoculum of several hundred cysts might be expected to be sufficient for a new intro-duction of this species (Hallegraeff 1998). Even if 99.9% of the original water werereplaced during the exchange, the number of cysts still remaining would be 5.5 x 107

(Rigby & Taylor 2001).

3.5 SAFETY ASPECTS

As long as safety permits, open–ocean ballast water exchange is currently the best op-tion available as a first step in order to minimize the number of species and individualsunintentionally introduced with ballast water until more effective treatment options areapproved. However, ballast water exchange can pose serious concerns and every even-tuality must be taken into account when deciding whether it would be safe to exchangeballast at sea.

A study of the "Ship Operational and Safety Aspects of the Ballast Water Exchange atSea" was carried out by Woodward et al. (1992) who concluded that ballast/deballastoperations may be carried out safely if wave heights were below a maximum value.Using hydrostatic data hull bending moments and stabilities are investigated to find thetank-emptying operations representing the maximum safety. At-sea analysis for hullbending moment, shear and rate of slamming was carried out using both linear andnon-linear analysis. From the small sample of three ships (a dry bulk carrier, a tankerand a container ship) it appears that the critical wave height lies between 10 and 20 feet.The sample is too small to support a more definite conclusion on the maximum safewave height. However, a complete reballasting at sea using the empty/refill method canbe unsafe even in good weather conditions owing to stability problems. A sequentialoperation of continuous flushing of tanks with ocean water would be an alternativeworkable option as it reduces stability problems (Rigby & Hallegraeff 1993).


Karaminas (2000) has analysed the use of sequential exchange (empty-refill) on 26existing ships of various types, configurations and sizes. This study showed that manyof the ships did not have sufficient strength to undertake this type of exchange safely, orin some cases, only under light or minimum ballast conditions. The use of a diagonalsequential method could be an effective method for reducing the still water bendingmoments and shear forces to within permissible levels. It is essential that any of theocean exchange options must only be undertaken if safe conditions can be maintained.This decision always rests with the Master of the ship.

In addition the IMO sub-committee on Ship Design and Equipment prepared guidelineson safety aspects of ballast water exchange taking into consideration:(i) crew safety,(ii) structural integrity as well as stability of ships and in particular stated to avoid

over and under pressurisation of ballast tanks,(iii) free surface effects on stability and sloshing loads in tanks that may be slack,(iv) admissible weather conditions,(v) maintenance of adequate intact stability in accordance with an approved trim and

stability booklet,(vi) permissible sea-going strength limits of shear forces and bending moments in

accordance with an approved loading manual,(vii) torsional forces,(viii) minimum/maximum forward and aft draughts, and(ix) wave induced hull vibration.

It was suggested that the ballast water management plan should include designatedcontrol personnel responsible for the ballast water exchange and crew training for fa-miliarisation. Furthermore, it was noted that a need exists to evaluate the safety of thelong term aspects of ballast water exchange, taking into account relevant safety matters,including safety of crews and ships, ship´s position, weather condition, ballast systeminspection and maintenance, machinery performance and availability (IMO MEPC39/7,IMO MEPC39/7/1, IMO MEPC39/7/4, IMO MEPC39/8).

From the above it can be summarised that every ship is different, even sister ships, inrelation to the safety of ballast water exchange carried out by the sequential or flowthrough methods. There are too many variables involved to make any specific conclu-sions suitable for all ships at all times by any regulating authority. If the owners or op-erators of a ship select ballast water exchange as their preferred option for ballast watermanagement, the safety of the ship and crew always resides with the master under inter-national law (SOLAS Convention). In addition, as the movement of ballast water at sea,within the hull envelope, of any vessel is considered a “CRITICAL SAFETY PROC-ESS” as it can affect the trim, stability, the bending moment or the shear forces actingupon the hull of the ship as well as the possibility of under or over pressurising the bal-last water tanks, hence the safety of the ship and its crew. Being a critical safety processit is also a requirement under the ISM Code to have a procedure to cover any movementof ballast water at sea. As the movement of ballast water can have an affect on the trim,stability, bending moment or shear forces acting upon the hull of the ship, the operationrequires approval by the ship’s Classification Society and by the Flag State Administra-tion.

A. Taylor et al.496

4 Mechanical removal of species in ballast water

Mechanical technologies are based on particle-size or specific weight to separate orremove organisms from the ballast water. To its advantage mechanical treatment doesnot generally produce any unwanted environmental side effect.

4.1 FILTRATION

Filtration of ballast water is one of the most environmentally sound methods. The over-all advantage of this method is its use during ballast water intake. The backwashing(filter cleaning) material may be returned immediately to the region from which theballast water was taken without any treatment.

Filter systems under consideration include self-cleaning backwashing filter systems,microfiltration and granular filtration. Woven mesh filters, made from synthetic fibres,are available as automatically self-cleaning units. They could potentially be retro-fittedto existing ships or incorporated into the design of new vessels. Automatic cleaning ofthe filters can be programmed either for specific time intervals or at specific pressuredifferences across the filters. This would involve stainless steel brush and suction scan-ner filter mechanisms for ‘coarse’ and ‘fine’ filters. It is claimed that the ‘coarse’ filterwould remove most of the larger zooplankton whilst the second in-line ‘fine’ filterwould remove most of the smaller zooplankton and much of the large and medium sizedphytoplankton. In early experiments filter capacities of 500-1,000 m³ per hour wereachieved at the various mesh sizes tested. As pump capacities in smaller vessels, e.g.container ships and cruise liners, are below this figure filtration would not slow downthe ballasting or de-ballasting operations. However, resizing of the pumps may be re-quired to cope with the increased filter resistance. Alternatively, with no modification, areduction in pump capacity and concomitant increase in ballasting time could result.

Recent investigations included test filters with screen sizes of 25, 50 and 100 µm and aflow rate of about 5,000 m³ per hour. The fine screens were protected from larger float-ing objects and organisms by the additional use of 5 mm screens prior to the water entryinto the system. Filters reached a 92-99% percent removal of larger zooplankton and 74-94% removal of smaller zooplankton and phytoplankton (Taylor & Rigby 2001;GloBallast Programme 2001). Bacteria attached to larger organisms and suspendedsolids were significantly reduced using the smaller filter (25 µm). However, total bacte-ria remain largely unaffected by filtration. It was suggested that the effectiveness offilter systems could be increased by the use of an additional technique such as heat orUV as secondary treatment (Cangelosi et. al. in print; Rigby & Taylor 2001).

The first full scale investigation of a ballast water treatment system, other than mid-ocean exchange and one form of heat treatment, is a filtration experiment currently inuse on the cruise liner M/S Regal Princess (P&O Princess Cruises, United States) with atotal ballast tank capacity of 4,213 m³.

4.2 CYCLONIC SEPARATION

Cyclonic separation has been proposed as a relatively simple and inexpensive way ofremoving larger particles and organisms from ballast water. Water and particulates enter


a separation unit tangentially, thus inducing a circular flow. The water is then drawnthrough tangential slots and accelerated into the separation chamber. Centrifugal actionejects particles heavier than the water to the perimeter of the separation chamber. Sus-pended solids gently drop along the perimeter and end up in the calm collection cham-ber of the separator. In addition to the removal of organisms, a large fraction of thesediment usually deposited in the ballast water tanks could be removed. Solids may beperiodically purged or continuously bled from the separator. A flow rate of up to about3,000 m³ per hour has been achieved. The pressure drop across this separator is rela-tively low (by comparison with filtration), so that the pumps already present on theships would generally be sufficient. It seems feasible that this system could be used onboard ships by incorporating the separator in a ballast tank recirculation system. How-ever, cyclonic separation of organisms with a specific gravity similar to that of sea wa-ter (such as jellyfish, chaetognaths and phytoplankton) is likely to be limited (Arm-strong 1997).

To increase the overall efficacy a UV-light unit could be applied as secondary treatmentafter the cyclonic separation.

4.3 SEDIMENTATION AND FLOTATION

Other separation treatment processes such as sedimentation and flotation have also beenconsidered in terms of their application to ballast water treatment. The former essen-tially increases the settling of material from the water column under gravitational forces,with or without the use of coagulant chemicals to assist sedimentation. The latter alsoentails the use of coagulants and the injection of fine air bubbles into a ‘flotation tank’.The bubbles attach themselves to coagulated organisms or particulates and float them tothe surface. Efficacy data are not available (Carlton et al. 1995).

4.4 PUMPING VELOCITY

The use of a high velocity ballast water pumps during ballast water intake and dischargecould minimize the survival rate of some macro-organisms due to mechanical damage(Woodward 1990; Carlton et al. 1995). The installation of additional units in order tocreate high velocity jets of water in ballast tanks or pipework would involve high costs.Only limited data is available on the efficacy of this method (Carlton et al. 1995), al-though some recent trials using a pipe section fitted with internal mixing elements re-ported some progress (GloBallast Programme 2001). However, the usual maximumvelocity in ballast water systems should not exceed 3 m s-1, because of erosion of theinternal surfaces of the ballast pipes especially in the areas of ‘bends’ in the piping sys-tem. High velocity treatment techniques within the ballast system are extremely difficultto install and it is expensive to replace ballast water piping in service what makes thisoption less favourable.

5 Physical inactivation of species in ballast water

Although a number of physical techniques have been investigated, the most promisingoptions currently available are heat treatment and Ultra Violet (UV) irradiation. Physi-cal treatment options for ballast water focus on the removal of organisms or on chang-

A. Taylor et al.498

ing the physical properties or hydrodynamic characteristics of the water to kill / inacti-vate the taxa present.

5.1 HEAT TREATMENT

Temperatures around 40 °C will kill or inactivate many organisms of concern frequentlyfound in ballast water. Heating of ballast water to temperatures of this order has beenproposed as an appropriate means to kill most toxic marine organisms (AQIS 1993). Forexample, exposure to temperatures of 36 to 38 °C over a period of 2 to 6 h was suffi-cient to kill zebra mussels in pipes. More recent data showed that 35 to 38 °C for a pe-riod of 4 to 5 h effectively killed vegetative cells of toxic and non-toxic dinoflagellates(Hallegraeff et al. 1997). With many organisms, the temperature required will generallybe lower for longer periods of heating (Bolch & Hallegraeff 1993, 1994). In a laboratorystudy Bolch and Hallegraeff (1993) demonstrated that short-term (30 to 90 s) exposureto temperatures above 40 °C were effective in killing cysts of the dinoflagellates Gym-nodinium catenatum and Alexandrium tamarense, whereas temperatures as low as 35 to38 °C were sufficient after 4 h heating. Time temperature relationships required to kill anumber of model organisms have been examined by Mountfort et al. (1999, 2000).These laboratory findings were subsequently confirmed in full scale shipboard trials,where the ship's pipework was modified to enable waste heat from the main enginecooling circuit to heat the water in one of the ballast tanks by flushing with the heatedwater (Rigby et al. 1997, 1999). In these trials, onboard microscopic observation ofheated water, showed that none of the zooplankton (mainly chaetognaths and copepods)and only very limited original phytoplankton (mainly dinoflagellates) survived the heattreatment. The original organisms were reduced to flocculent amorphous detritus. Sub-sequent culturing efforts on the heated ballast tank samples only produced growth ofsome small (5µm) diatoms and colourless ciliates. Although no toxic dinoflagellatecysts were present in the tanks, based on earlier laboratory experiments, it is probablethat these would have been effectively killed by the temperatures achieved during theheating trial, since essentially all of the water reached 37-38°C.

Heating of ballast water as described above also has the added advantage that organismscontained in sediments would also be subjected to these temperatures (in fact highertemperatures are experienced at the bottom of the tanks where the ballast water ispumped into the base of the tanks).

This form of heating may not be appropriate for short (domestic) voyages or where heatlosses to the ocean are high (for example where sea temperatures are low).

In addition to the heating/flushing option, a number of other options (involving recircu-lation of water from the ballast tanks) utilising alternative heat exchanger designs and/oradditional heat or steam from the main engine or exhaust system are possible (Sobol etal. 1995; Thornton 2000; Mountfort et al. 1999, 2000; Rigby & Taylor 2001). The suit-ability, cost and practicality of these alternatives will depend on the ship's heat balance,length of voyage, engine cooling system design, availability of steam and the additionalequipment involved.


Although some concerns have been expressed about possible effects of elevated tem-peratures on tank internal coatings (Armstrong 1997), these effects are not likely to besignificant given the temperatures involved and time that the tank surfaces are exposed(Rigby et al. 1997).

Concerns that such mild heating of ballast water could stimulate the growth of patho-genic bacteria such as Vibrio cholerae have not been substantiated by simulated labo-ratory culture experiments (Desmarchelier & Wong 1998).

5.2 COOLING TREATMENT

A reduction of the temperature of the ballast water near the freezing point requires e.g. acooling unit, additional pipework and power, not to mention the safety aspects relatingto the ships hull and tanks that could occasion if the water were to freeze. Further stud-ies to evaluate the feasibility should take into account the safety aspects, likely costs,temperature related impact on the pipework and ballast tanks as well as trials on thetreatment effectiveness on ballast water organisms.

5.3 ULTRAVIOLET IRRADIATION

UV irradiation is commonly used for sterilising large amounts of potable or wastewaterand for the purification in aquaculture and fisheries (Carlton et al. 1995). UV irradiationoperates by causing photochemical reactions of biological components such as nucleicacid (DNA and RNA) and proteins. The lower UV wavelengths are generally moreeffective. However, irradiation at these wavelengths shows a lower transmission inwater. Due to a higher concentration of inorganic solutes, the transmission in seawateris slightly less than in freshwater. It may further be affected by the organic load, sus-pended solids or air bubbles. The effectiveness of UV treatment further depends largelyupon the size and morphology of organisms. Viruses require similar dosages to bacteria.Algae require larger dosages than bacteria due to their size and pigmentation.

The biological effectiveness of UV treatment is not necessarily a simple function ofirradiance and exposure time. Experiments with a phytoplankton species showed that ashort exposure at high irradiance was found to be more effective than long exposure atlow irradiance (Cullen & Lesser 1991). Confirmation by Hallegraeff et al. (1997) andMontani et al. (1995) have shown that the germination of cysts of Alexandrium, Gym-nodinium, Protoperidinium, Scrippsiella and Gyrodinium occurred after exposure to UVradiation. However, it has to be noted that many cyst walls are impermeable to UV. Ithas been suggested that organisms have repair mechanisms which may enable them torecover from the UV treatment provided their exposure does not exceed a certain time.Similar observations were made with bacteria and other phytoplankton species, demon-strating the importance of specifying exposure time as well as irradiance level.

In-line flow treatment would appear feasible and the most practical option forretro-fitting a UV treatment system on ships. Treatment could take place at the time ofballasting and/or de-ballasting. This method is likely to be practicable and environmen-tally sound (no toxic side effects) and no adverse effects in pipework pumps, sealingrings or coating are known (Müller 1995; Müller & Reynolds 1995). No specific health,safety or environmental concerns appeared to be associated with the use of UV systems

A. Taylor et al.500

on board ship, however, the possibility exists that UV radiation might cause mutation ofgenetic material in the organisms treated. Capital and running costs for suitable systemsare likely to be significant (Müller 1995; Müller & Reynolds 1995; Rigby & Taylor2001).

Disadvantages of UV treatment include the possibility that some smaller organismscould pass the UV unit in the shadow of larger organisms or suspended solids withoutany treatment (Armstrong 1997), the reduced penetration of UV irradiation in turbidballast waters, what is a major limitation for the use in ballast water treatment (Rigby etal. 1993) and the recovery of the phytoplankton following exposure to UV irradiation.On the basis of the experimental results, UV appears promising for the treatment ofsome organisms in ballast water and its effectiveness may be increased by initial filter-ing of ballast water to avoid shadowing effects of larger organisms and suspended sol-ids.

5.4 ULTRASONICS

The use of ultrasonics for controlling hull fouling dates back to the 1950s, however,demonstration of its potential application for ballast water treatment purposes has notbeen investigated to any great extent (Subklew 1963; Müller 1995; Müller & Reynolds1995). Ultrasound is thought to be mediated through various responses that may be fatalto marine organisms. These are heat generation, pressure wave deflections, cavitationand possibly the degassing effect of ultrasound causing removal of much of the oxygen.Higher frequencies, warmer temperatures and lower concentrations of dissolved matterhave been found to increase the effect of ultrasound pulses. Plankton mortality has alsobeen observed in the presence of ultrasound and is considered in part to be attributableto the cavitation process (Armstrong 1997).

The implementation of ultrasonics would require the installation of in-line transducersbecause ultrasound is unlikely to penetrate sediments. With respect to health and safetyaspects, problems may arise with noise from some transducer types. There may also besome as yet unknown implications for the ship’s structural integrity and health of crewsfollowing repeated exposure to ultrasound. It has also been suggested that the cavitationprocess could cause damage to tank coatings or structures (Müller 1995; Müller & Rey-nolds 1995). The application of sonic disruption as treatment to the ballast water has sofar not been tested.

5.5 MICROWAVE

Beside the use of ultrasonics the application of microwaves (wavelength 0.1-1 nm) forballast water treatment has been listed. The application of microwaves is relatively newbut has been developed to treat waste waters (AQIS 1993). Microwaves have not beentested to treat ballast water organisms.

5.6 ELECTRICAL REMOVAL OF SPECIES IN BALLAST WATER

Ballast water treatment with electrical currents may cause serious damage to macro-organisms (Woodward 1990). The inactivation of dinoflagellate cysts had beenachieved by the use of an electric shock during the exposure to 100V for 5 sec (Montani


et al. 1995). Hallegraeff et al. (1997) claim that this mortality is due to generation ofheat and reactive oxygen species.

The installation of an electronic unit in the sea chest of the vessel near the ballast waterintake may provide a deterrent for the entry of some macro-organisms. Comparableunits have been used in intake areas of power plants to prevent the unwanted impact ofmacro-organisms but have not been implemented onboard ships.

5.7 ELECTROCHEMICAL BALLAST WATER TREATMENT

An electrochemical technology using large-scale Boron-Doped Diamond (BDD) elec-trodes (electrolysis cell) was recently tested in laboratory trials. The self-cleaning BDDelectrodes show significant technical and economic advantages due to their specificproduction capacity of disinfectant and oxidant substances. Completed treatment trialsshowed that the system has an outstanding production of chemical oxidizing agentssuch as strong hydroxyl radicals. In addition it may also support the production of otheroxidizing agents like chlorine, persulfates, ozone and hydrogen peroxide in many kindsof water. The disinfectants produced by the system are of very high activity (i.e. morethan four times higher than a typical chemical dosage at similar concentrations withregard to total chlorine). The flow rate of 7,000 m3 ballast water in two days resulted ina total energy consumption <420 kWh (pumping included). The system needs littleextra maintenance and energy compared to other treatment techniques, has a compactdesign (enabling retrofitting onboard existing ships) and is highly effective to treat mi-croorganisms being of major concern in unwanted ballast water introductions (e.g. dis-ease agents and pathogens) (Pupunat et al. 2001, CSEM, Switzerland and Tanksystem,Switzerland pers. comm.).

5.8 GAS SUPER-SATURATION

As a coarse definition gas supersaturation (GSS) is a condition where the water containsmore dissolved gas than it normally will at a given depth and barometric pressure.When GSS is established, it will tend to move towards equilibrium.

Several groups of organisms have been found to be sensitive to GSS and when exposedto high levels of GSS, they may suffer from acute Gas Bubble Trauma (Colt 1986).Acute gas bubble trauma means formation of bubbles in tissues and the vascular system,which often leads to occlusions of vessels, emphysema, and haemorrhages (Weitkamp& Katz 1980). Jelmert (1999) has suggested that this susceptibility could be utilized asthe target for an environmentally friendly ballast water treatment, where air (approxi-mately 80% N2) or N2 are injected into the ballast water.

While only multicellular organisms are believed to be directly susceptible, some addi-tional effects can be hypothesised. The use of N2 will reduce the available O2 contentand impair the respiration of all aerobic organisms. The injection of large amounts ofgas by a dedicated injection system will produce a high number of small bubbles. Theserepresent a large hydrophobic area sifting through the water column. Inorganic particles,as well as bacteria and other organisms with hydrophobic surfaces will be transported tothe surface. Organisms with cavities or structures where small bubbles may be en-trapped will also be transported to the surface where an increased bacterial activity

A. Taylor et al.502

could increase the breakdown of injured or impaired organisms. The method is undertesting and evaluation. However, at high levels of supersaturation lethal treatment hasbeen demonstrated (Jelmert pers. comm.).

6 Chemical and biocidal inactivation of species in ballast water

A large number of chemical disinfectants are commercially available that have beenused successfully for many years in land-based potable and wastewater treatment appli-cations. Target organisms include protozoa, vegetative and resting cells of algae, bacte-ria and viruses. Within the general treatment efficacy requirements, that each chemicaltreatment of ballast water needs to meet, it should be environmentally sound and anyresidual chemical should be fully biodegradable to avoid accumulation of toxic sub-stances in the remaining ballast water and/or sediments inside the ballast water tanks.

The costs involved in the use of some chemicals, operating and material costs are con-sidered to be prohibitive (Müller 1995; Müller & Reynolds 1995; Rigby et al. 1993;Rigby & Taylor 2001). Beside the costs, the storage and use of chemicals on boardcould be impracticable.

With some chemicals several tonnes would be required to treat the large amounts ofballast water on large ships such as bulk liquid and ore carriers being fully loaded withballast water. In addition, both inorganic and organic biocides would present a range ofhealth and safety problems related to the storage and handling of chemicals, their com-patibility with cargoes carried onboard ships, as well as those related to the direct andindirect handling of chemicals by ships' crews.

Most of the traditional biocides produce by-products which are likely to be environ-mentally unacceptable and/or might require specialist operator skills. Suitable dispersalmechanisms would also need to be addressed, in particular mechanisms for penetratingthe sediment layer. The injection during ballast water uptake seems to be an appropriateoption to apply and dose the chemical. In cases where ballast water in cargo holdswould be treated with chemicals, these tanks would have to be cleaned intensively be-fore cargo could be loaded in the same tanks (Carlton et al. 1995). Although there isextensive testing ongoing for possible residues, the long-term accumulating effects areunknown.

Biocides suggested for the use to treat ballast water include ozone, hydrogen peroxide,potassium permanganate, chlorination (chlorine dioxide, chloramines and so-dium/calcium hypochlorite), ozonation, electrolytically generated metal ions (copperand silver ions), oxygen deprivation (de-oxygenation) using e.g. reducing agents, suchas sulphur dioxide or sodium sulphite, coagulants (aluminium and ferric sulphate, ferricand poly aluminium chloride as well as cationic polymers), pH and salinity adjustment,antifouling paints as ballast tank coatings, organic biocides (3-trifluoromethyl-4-nitophenol, bromine, formaldehyde, glutaraldehyde, isothiazolone, quaternary ammoniacompounds, sodium amines, EDTA, peracetic acid, bisulfite, iodine (periodate), use ofbacterial pyrogens (endotoxins) and organic algacides containing isothiazolone) (Ridg-way & Safarik 1991; Bolch & Hallegraeff 1994; Voigt 2000; Fuchs (Degussa) pers.comm.; GloBallast Programme 2001).


The most recently proposed treatment technique to inactivate or kill microoganisms inballast water using chemicals is an advanced oxidation technology. This patented tech-nology consists of a combination of ozone (O3), two UV systems with different wave-length spectra and catalysts. The suggested ozonolytic/photolytic/photocatalytic redoxprocesses are stated to operated simultaneously in a closed system consisting of a reac-tor and a control panel. The reactor is made of titanium and houses low-pressure UVlamps and catalysts placed close to the lamps. A large number of radicals are generatedwithin the reactor. These hydroxyl radicals are aggressive and can break down virtuallyany organic compound to carbon dioxide and water. They are reported to be short lived(a few nanoseconds). This technique could be considered as an effective secondarytreatment after particles and large organisms are removed by e.g. filter systems. Thefirst prototype of the system will have a flow rate of several hundred m³ per hour, whilesystems with further increased capacity are under development. Performance or costdata are not available at present (Patrick Dahl, pers. comm. BenRad Marine TechnologyAB, Stockholm Sweden). Possible corrosion problems (production of aggressive hy-droxyl radicals) need to be considered unless the reactor is well separated from the mainpiping system.

One advantage of most chemicals is the comparatively easy application to the ballastwater when it is taken on board. However, one of the shortcomings is that with every(even with partly) exchange of ballast water, the entire volume of water in the ballasttank has to be treated. Otherwise the dilution of the substances with the remaining waterwould result in insufficient dosing of the chemicals for the newly added ballast water.This limits the applicability of stand-alone chemical treatments to those types of ships,that exchange the ballast water completely in one lot (e.g. bulk carriers and oil tankers).In contrast, container vessels frequently exchange only part of their ballast water, ortransfer part of the ballast water between ballast tanks when loading or unloading in theports of call. This problem can be overcome if all the ballast water on board is treated tothe desired level, however when ballast water is loaded on top of existing ballast waterit would require the whole tank to be retreated with the chemical.

7 Constant volume of ballast water

Another option is the carriage of a constant volume of ballast water onboard withoutany discharges or uptakes of additional water. This option seems to be applicable to avery limited number of vessels. Ships which usually carry very little amounts of ballastwater as e.g. cruise liners, could minimize their ballast water discharge to a minimum oreven could prevent any discharge by pumping the ballast on board from one tank intoanother.

8 Alternating salinities in ballast water and area of discharge

Wherever possible, alternating salinities of ballast water and area of discharge for bal-last water could be used. Firstly, to discharge marine ballast water in freshwater areas(e.g. the North American Great Lakes, freshwater ports) and secondly, to dischargefreshwater ballast in marine ports could help to minimize the survival of organisms afterdischarge. It is believed that most of the freshwater organisms cannot survive marineconditions and vice versa. Knowing that there are exceptions to this rule, this option

A. Taylor et al.504

could be used to minimize the risk but cannot exclude further species introductions.However, there are many trading routes in the world where ships do not have the op-portunity to take freshwater ballast onboard. The use of an onboard desalination unit forthis purpose is probably extremely time consuming and associated with enormous needof energy.

9 Fresh or treated water ballasting

Providing ships with treated or fresh water appears to be a useful option in unique cir-cumstances (Carlton et al. 1995; Rigby & Taylor 2001). An International Seminar onFresh Water Ballasting in 1983 discussed the use of fresh water ballast for oil carriers. Itwas suggested that oil carriers could load fresh water instead of oceanic ballast waterbefore cruising back to the oil exporting country without any cargo. Many of the oilexporting countries are located in arid or semi-arid climates where rainfall is scarce. Theagriculture of some countries could benefit from this fresh water imported in segregatedballast tanks. The applicability of this option is limited to very special circumstances oncertain trading routes and the availability of freshwater at donor ports.

10 Reception facilities

The possibility of land-based reception facilities for ballast water has not been ruled outfor the treatment of smaller volumes of ballast water. The reception facility would needto be placed in certain areas of ports taken into account quarantine regulations. Land-based facilities could provide an acceptable means of control, but appear to have veryhigh costs involved e.g. for pipework of large diameter in larger ports. Furthermoreships would need to be equipped with appropriate piping systems to connect the ballastwater outlet with the ports ballast water pipeline. Many ports do not have any land areasavailable to house such storage facilities.

In order to reduce costs for the pipework an especially designed tanker (ballast watertreatment vessel) could act as reception facility. The ballast water to be dischargedcould be pumped to this vessel located along side the discharging vessel (AQIS 1993;Taylor & Rigby 2001).

11 Cost effectiveness of various treatment options

In addition to the technical effectiveness of the various treatment options, the cost ef-fectiveness will play an important role in the selection and long-term viability of a par-ticular treatment option (Rigby & Taylor 2001; Table 3).

Exchange of ocean water in its simplest form (with no additional equipment) providesthe most cost effective option (1.48-2.25c/m3). These costs are reduced by approxi-mately 50% (for the empty/refill option) if gravity ballasting can be accomplished. Thecapital costs associated with additional equipment (that could be required in some casesfor safe or effective operation) can result in an increase up to approximately 18.67c/m3 .The heating /flushing process provides the next most cost effective option at 3.35c/m3.Use of a Hi Tech Marine Systems (Thornton 2000) involving recirculation, higher tem-peratures and additional heat exchange equipment has been estimated at = 5.42c/m3. A


range of representative costs for a selection of ships and other potential treatment op-tions are also summarised in Table 3.

Costs for the container ship are relatively high compared to the other ships (where addi-tional capital equipment is required) since the quantity of water treated is quite low. Itmay be possible to reduce some of these capital costs by reducing the capacity of thenew equipment. However this aspect would need to be considered as part of the devel-opment of the Ballast Water Management Plan to allow the optimum overall outcome tobe achieved.

Table 3. Indicative comparative ballast water treatment costs. [Costs have been esti-mated for three different types of ship classes; a Capesize Bulk Carrier, an LNG Carrierand a Container Vessel after Rigby & Taylor (2001). Capital costs have been based onthe use of a Capital Recovery Factor (incorporating an 8% interest rate over a 10 yearperiod) of 0.149 and a voyage schedule generally involving ballasting 12 times per year.Operating costs have been based on the use of additional fuel and additional mainte-nance associated with operating the equipment. The estimated costs are based on thevolume of ballast water on board the ship, except in the case of the container vesselwhere a cost is indicated for the quantity of ballast water actually replaced or treated, ascontainer ships do not usually completely discharge their ballast water in port]. Efficiencydetails regarding removal or inactivation of species see text.

Treatment option Costs per m³[US$]

Exchange of ocean water in its simplest form (with no additional equip-ment).Costs are reduced by approximately 50% (for the empty/refill option) ifgravity ballasting can be accomplished

1.48c-2.25c

Exchange of ocean water including capital costs associated with additionalequipment (that could be required in some cases for safe or effectiveoperation)

up to 18.67c

Heating/flushing process 3.35c

Using recycled process water 4.16c

Heating/flushing process using a Hi Tech system (Thornton 2000) involv-ing recirculation, higher temperatures and additional heat exchangeequipment

5.42c

Hydrocyclones 6.49c-26.33c

Continuous backflushing filtration (with a relatively high capital cost com-ponent)

7.07c-19.31c

UV irradiation 9.80c-31.39c

Chemical treatment (based on operating cost alone) 14.46c-$24.10UV combined with hydrocyclone 16.28c-57.71c

UV combined with filtration 16.87c-50.69c

Land based treatment 20.48c-$8.31Dedicated treatment ship 32.53c

Use of fresh water 50.00c-72.29c

The relatively high costs of some options (resulting from the equipment capital costs)will probably mean that preference will be given to those involving little or no capital.However standards which take into account biological effectiveness will ultimately havean influence on the most appropriate choice. It is important to note that the shippingindustry has currently generally accepted the costs of ballast water exchange (as speci-fied in the current IMO Guidelines) as being reasonable. Treatment technologies in-

A. Taylor et al.506

volving significantly higher costs will have a direct impact on freight rates. Further, ithas to be noted that the most cost effective method is not necessarily at the same timethe most effective method regarding its removal or inactivation capacity of species inballast water.

The capital cost accounts for a large proportion of the overall cost of retrofitting equip-ment to existing ships for some treatment options. This situation will be less of an issuein relation to a new ship where new designs can be readily included and the additionalcapital may be very minor compared to the total cost of the ship.

12 Improved designs to enhance better ballast water management and treatmenton new and existing ships

The adoption of many of the above treatment or management options will require theretrofitting or modification of existing pipework or equipment to permit the new proce-dures to be put into practice in a safe, technically effective, environmentally acceptable,practical and cost effective way. For many ships this will involve substantial costs.

For new ships the cost of incorporating new designs and installation of new equipmentwill represent a very minor additional cost. Consequently it is important that adequateconsideration be given to these concepts at the new ship design stage.

The following design concepts have been suggested to facilitate and enhance betterballast water management and treatment. Several of these are specifically related tofacilitating the implementation and operation of the various forms of ocean exchange,water drainage and minimisation of sediment accumulation as well as water samplingrequired for research, monitoring and compliance testing (more specific details havebeen presented by Taylor & Rigby 2001):

12.1 NEW AND EXISTING SHIPS

(i) fitting of tanker hatches, where possible, as an alternative to manholes to allowmore ready access to tanks would be beneficial. It is also suggested that the tankimmediately below the opening be kept free of obstructions that may impedelowering of sampling nets and other equipment

(ii) fitting of quick-release couplings (such as Kamlock Coupling Caps) to soundingpipes or sampling pipes would be beneficial as well as their specific locationwithin ballast tanks to enhance sampling

(iii) modifications to standard sounding pipes to allow for better sampling by incorpo-rating a number of holes located circumferentially down the length of the pipe(say 25 mm diameter, 1 m apart) to allow for relatively free flow of water in thetank to be represented and to make sampling at a particular position more effec-tive

(iv) new ships' design attention to providing access to tanks (especially where accessis not normally required) to enhance sampling of sediments

(v) if ocean exchange using the empty/refill option is selected, consideration shouldbe given at the design stage to have sufficient strength built into the hull girder toallow this operation to be undertaken safely


(vi) if ocean exchange using the continuous flushing option is selected, new shipdesigns should examine options to allow ballast water overflow to take place in asafe and convenient manner; examples include- doubling the number air pipes- installation of Tanker Hatches; and- installation of internal overflow pipes to avoid water flowing over the deck

(vii) if ocean exchange using the dilution method is selected, consideration can begiven to the installation of deck pipes to load ballast water through the top of theballast tanks (IMO MEPC 38/13/2 1996) or alternative internal piping arrange-ments (Armstrong 1997)

(viii) if heating and heating /flushing is selected as the treatment option (Rigby &Taylor 2001) it is recommended that appropriate pumps and piping be includedin new ship designs. Heating could also be enhanced by designing the ballasttanks so that there is provision for one tank to be empty at any one time and thateach tank is strengthened to allow ballast water to be pumped sequentiallythrough a heating system and then into an empty tank. Under these conditions,treatment time would be minimised and the problem of mixing of partiallytreated water with treated water would be avoided. This arrangement would alsorequire an additional pump for circulation and pipework to permit water to betransferred in/out of all tanks

12.2 BALLAST TANK DESIGN FOR NEW SHIPS

(i) install additional drain holes in longitudinals and intercostals(ii) install larger drain holes in floors at intersection of longitudinals and intercostals;

and(iii) install larger drain holes in horizontal and vertical longitudinals, corner gussets,

panting stringers and intercostals where they butt up against watertight bulkheadsto stop “hang up” of sediment and particles. This will allow better flow into theballast water suction and stripping heads located in the last aft bays of the tanks.Naval Architects should undertake flow calculations (CFD) to determine the sizeof the drain holes to confirm that there is sufficient flow of water to the suctionheads to match the capacity of the discharge of the ballast system. This will en-sure that the maximum amount of water and sediment is discharged with theminimum ”hang up” or retention of sediment in the tanks.

(iv) where possible ”Hat Box” suctions be installed in tanks and ballast water holds(v) ballast tanks should be installed with Butterworth type tank washing systems to

minimise the retention of sediments. This suggestion is good for existing andnew ships and in addition at the design stage consideration should be given to theenhanced manual removal of sediments at dry dockings.

Acknowledgements

The approval of the Department of Agriculture, Fisheries and Forestry - Australia toinclude some information from the Ballast Water Research Series Reports No. 12 and13 is acknowledged with thanks.

PREVENTIVE TREATMENT AND CONTROL TECHNIQUES … · preventive treatment and control techniques for...

Documents

Transcript of PREVENTIVE TREATMENT AND CONTROL TECHNIQUES … · preventive treatment and control techniques for...